I spend a lot of time trying to recreate functional test failures on
the command line where I can add gdbstub or run under rr. Invariably
this involves running the test with "env QEMU_TEST_KEEP_SCRATCH=1" and
then extracting the command line and tweaking it.

I finally got annoyed enough to look at the command line code and see
if I could do this a better way. Ideally I could just do something
like:

  ./pyvenv/bin/meson test --suite thorough func-arm-aspeed_anacapa --cmdline

but for now I'll stick with copy and pasting a line from the logs and
saving a few seconds removing all the QMP and console sockets.

I've also included the pylint change which is now reviewed and can be
picked up. If not I'll grab it next time I do a testing/next sweep.

Alex.

Alex Bennée (4):
  python/qemu: split arg between base and harness lists
  python/qemu: split console from harness args
  python/qemu: dump a developer friendly version of cmdline to logs
  tests/functional: tell pylint not to check c-modules

 python/qemu/machine/machine.py | 29 ++++++++++++++++++++++++-----
 tests/functional/pylintrc      |  1 +
 2 files changed, 25 insertions(+), 5 deletions(-)

-- 
2.47.3

Alex Bennée <alex.bennee@linaro.org> writes:

> I spend a lot of time trying to recreate functional test failures on
> the command line where I can add gdbstub or run under rr. Invariably
> this involves running the test with "env QEMU_TEST_KEEP_SCRATCH=1" and
> then extracting the command line and tweaking it.
>
> I finally got annoyed enough to look at the command line code and see
> if I could do this a better way. Ideally I could just do something
> like:
>
>   ./pyvenv/bin/meson test --suite thorough func-arm-aspeed_anacapa --cmdline
>
> but for now I'll stick with copy and pasting a line from the logs and
> saving a few seconds removing all the QMP and console sockets.
>
> I've also included the pylint change which is now reviewed and can be
> picked up. If not I'll grab it next time I do a testing/next sweep.
>

Queued to testing/next, thanks.

-- 
Alex Bennée
Virtualisation Tech Lead @ Linaro

On 14/06/2026 12.08, Alex Bennée wrote:
> I spend a lot of time trying to recreate functional test failures on
> the command line where I can add gdbstub or run under rr. Invariably
> this involves running the test with "env QEMU_TEST_KEEP_SCRATCH=1" and
> then extracting the command line and tweaking it.
> 
> I finally got annoyed enough to look at the command line code and see
> if I could do this a better way. Ideally I could just do something
> like:
> 
>    ./pyvenv/bin/meson test --suite thorough func-arm-aspeed_anacapa --cmdline
> 
> but for now I'll stick with copy and pasting a line from the logs and
> saving a few seconds removing all the QMP and console sockets.
> 
> I've also included the pylint change which is now reviewed and can be
> picked up. If not I'll grab it next time I do a testing/next sweep.

Series
Reviewed-by: Thomas Huth <thuth@redhat.com>

Feel free to pick it up through your testing tree, I don't have much other 
things currently pending that would justify a pull request from my side 
right now.

  Thanks,
   Thomas